UNet-VITS: Elevating Single-Stage TTS Quality with Spectral Restoration and Post-Processing Optimization
This work proposes an enhanced VITS2 architecture, UNet-VITS, to address these long-standing challenges through three synergistic technical improvements—a unique combination not seen in prior research.
Min Zheng, Danqing Liu, Tengyue Yang et al.
· 0 citations